Face Generation

In this project, you'll use generative adversarial networks to generate new images of faces.

Get the Data

You'll be using two datasets in this project:

  • MNIST
  • CelebA

Since the celebA dataset is complex and you're doing GANs in a project for the first time, we want you to test your neural network on MNIST before CelebA. Running the GANs on MNIST will allow you to see how well your model trains sooner.

If you're using FloydHub, set data_dir to "/input" and use the FloydHub data ID "R5KrjnANiKVhLWAkpXhNBe".

In [1]:
#data_dir = './data'
#FloydHub - Use with data ID "R5KrjnANiKVhLWAkpXhNBe"
data_dir = '/input'


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import helper

helper.download_extract('mnist', data_dir)
helper.download_extract('celeba', data_dir)
Downloading mnist: 9.92MB [00:07, 1.33MB/s]                            
Extracting mnist: 100%|██████████| 60.0K/60.0K [00:11<00:00, 5.39KFile/s]
Downloading celeba: 1.44GB [00:24, 58.9MB/s]                               
Extracting celeba...

Explore the Data

MNIST

As you're aware, the MNIST dataset contains images of handwritten digits. You can view the first number of examples by changing show_n_images.

In [2]:
show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
%matplotlib inline
import os
from glob import glob
from matplotlib import pyplot

mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'mnist/*.jpg'))[:show_n_images], 28, 28, 'L')
pyplot.imshow(helper.images_square_grid(mnist_images, 'L'), cmap='gray')
Out[2]:
<matplotlib.image.AxesImage at 0x7fc838ece048>

CelebA

The CelebFaces Attributes Dataset (CelebA) dataset contains over 200,000 celebrity images with annotations. Since you're going to be generating faces, you won't need the annotations. You can view the first number of examples by changing show_n_images.

In [3]:
show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'img_align_celeba/*.jpg'))[:show_n_images], 28, 28, 'RGB')
pyplot.imshow(helper.images_square_grid(mnist_images, 'RGB'))
Out[3]:
<matplotlib.image.AxesImage at 0x7fc83add8a90>

Preprocess the Data

Since the project's main focus is on building the GANs, we'll preprocess the data for you. The values of the MNIST and CelebA dataset will be in the range of -0.5 to 0.5 of 28x28 dimensional images. The CelebA images will be cropped to remove parts of the image that don't include a face, then resized down to 28x28.

The MNIST images are black and white images with a single color channel while the CelebA images have 3 color channels (RGB color channel).

Build the Neural Network

You'll build the components necessary to build a GANs by implementing the following functions below:

  • model_inputs
  • discriminator
  • generator
  • model_loss
  • model_opt
  • train

Check the Version of TensorFlow and Access to GPU

This will check to make sure you have the correct version of TensorFlow and access to a GPU

In [4]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
from distutils.version import LooseVersion
import warnings
import tensorflow as tf

# Check TensorFlow Version
assert LooseVersion(tf.__version__) >= LooseVersion('1.0'), 'Please use TensorFlow version 1.0 or newer.  You are using {}'.format(tf.__version__)
print('TensorFlow Version: {}'.format(tf.__version__))

# Check for a GPU
if not tf.test.gpu_device_name():
    warnings.warn('No GPU found. Please use a GPU to train your neural network.')
else:
    print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))
TensorFlow Version: 1.1.0
Default GPU Device: /gpu:0

Input

Implement the model_inputs function to create TF Placeholders for the Neural Network. It should create the following placeholders:

  • Real input images placeholder with rank 4 using image_width, image_height, and image_channels.
  • Z input placeholder with rank 2 using z_dim.
  • Learning rate placeholder with rank 0.

Return the placeholders in the following the tuple (tensor of real input images, tensor of z data)

In [5]:
import problem_unittests as tests

def model_inputs(image_width, image_height, image_channels, z_dim):
    """
    Create the model inputs
    :param image_width: The input image width
    :param image_height: The input image height
    :param image_channels: The number of image channels
    :param z_dim: The dimension of Z
    :return: Tuple of (tensor of real input images, tensor of z data, learning rate)
    """
    inputs_real = tf.placeholder(tf.float32, [None, 
                                              image_width, 
                                              image_height, 
                                              image_channels], 
                                              name='inputs_real')
    
    inputs_z = tf.placeholder(tf.float32, [None, z_dim], name='inputs_z')
    lr = tf.placeholder(tf.float32)
    
    return inputs_real, inputs_z, lr


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_inputs(model_inputs)
Tests Passed

Discriminator

Implement discriminator to create a discriminator neural network that discriminates on images. This function should be able to reuse the variables in the neural network. Use tf.variable_scope with a scope name of "discriminator" to allow the variables to be reused. The function should return a tuple of (tensor output of the discriminator, tensor logits of the discriminator).

In [8]:
def discriminator(images, reuse=False):
    """
    Create the discriminator network
    :param images: Tensor of input image(s)
    :param reuse: Boolean if the weights should be reused
    :return: Tuple of (tensor output of the discriminator, tensor logits of the discriminator)
    """
    #real_image 28X28X3
    alpha = 0.2
    with tf.variable_scope('discriminator', reuse=reuse):
        
        conv_1 = tf.layers.conv2d(images, 64, 5, strides=2, padding='same')
        lrelu_1 = tf.maximum(alpha * conv_1, conv_1)
        #14X14X64 now
        
        conv_2 = tf.layers.conv2d(lrelu_1, 128, 5, strides=2, padding="same")
        bn2 = tf.layers.batch_normalization(conv_2, training=True)
        lrelu_2 = tf.maximum(alpha * bn2, bn2)
        #7X7X128 now

        conv_3= tf.layers.conv2d(lrelu_2, 256, 5, strides=2, padding="same")
        bn3 = tf.layers.batch_normalization(conv_3, training=True)
        lrelu_3= tf.maximum(alpha * bn3, bn3)
        #4X4X256 now
        
        flat= tf.reshape(lrelu_3, (-1, 4*4*256))

        logits = tf.layers.dense(flat, 1, activation=None)
        out = tf.sigmoid(logits)

    return out, logits


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_discriminator(discriminator, tf)
Tests Passed

Generator

Implement generator to generate an image using z. This function should be able to reuse the variables in the neural network. Use tf.variable_scope with a scope name of "generator" to allow the variables to be reused. The function should return the generated 28 x 28 x out_channel_dim images.

In [9]:
def generator(z, out_channel_dim, is_train=True):
    """
    Create the generator network
    :param z: Input z
    :param out_channel_dim: The number of channels in the output image
    :param is_train: Boolean if generator is being used for training
    :return: The tensor output of the generator
    """
    reuse = not is_train
    alpha = 0.2
    with tf.variable_scope('generator', reuse=reuse):
        
        net = tf.layers.dense(z, 4*4*512)        
        net = tf.reshape(net, (-1,4,4,512))
        net = tf.layers.batch_normalization(net, training=is_train)
        net = tf.maximum(alpha * net, net)
        #4X4X512 
        
        net = tf.layers.conv2d_transpose(net, 256, 4, strides=1, padding="valid")
        net = tf.layers.batch_normalization(net, training=is_train)
        net = tf.maximum(alpha * net, net)
        #7X7X256
        
        net = tf.layers.conv2d_transpose(net, 128, 4, strides=2, padding="same")
        net = tf.layers.batch_normalization(net, training=is_train)
        net = tf.maximum(alpha * net, net)
        #14X14X128
        
        logits = tf.layers.conv2d_transpose(net, out_channel_dim, 4, strides=2, padding="same")
        out = tf.tanh(logits)
        #28X28X5
        
    return out


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_generator(generator, tf)
Tests Passed

Loss

Implement model_loss to build the GANs for training and calculate the loss. The function should return a tuple of (discriminator loss, generator loss). Use the following functions you implemented:

  • discriminator(images, reuse=False)
  • generator(z, out_channel_dim, is_train=True)
In [10]:
def model_loss(input_real, input_z, out_channel_dim):
    """
    Get the loss for the discriminator and generator
    :param input_real: Images from the real dataset
    :param input_z: Z input
    :param out_channel_dim: The number of channels in the output image
    :return: A tuple of (discriminator loss, generator loss)
    """
    d_model_real, d_logits_real = discriminator(input_real, reuse=False)
    g_model = generator(input_z, out_channel_dim, is_train=True)
    d_model_fake, d_logits_fake = discriminator(g_model, reuse=True)
    
    d_loss_real = tf.reduce_mean(
                      tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_real, 
                                                              labels=tf.ones_like(d_logits_real) * (0.9)))
    d_loss_fake = tf.reduce_mean(
                      tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, 
                                                              labels=tf.zeros_like(d_logits_fake)))
    d_loss = d_loss_real + d_loss_fake

    g_loss = tf.reduce_mean(
                 tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake,
                                                         labels=tf.ones_like(d_logits_fake)))
    
    return d_loss, g_loss


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_loss(model_loss)
Tests Passed

Optimization

Implement model_opt to create the optimization operations for the GANs. Use tf.trainable_variables to get all the trainable variables. Filter the variables with names that are in the discriminator and generator scope names. The function should return a tuple of (discriminator training operation, generator training operation).

In [11]:
def model_opt(d_loss, g_loss, learning_rate, beta1):
    """
    Get optimization operations
    :param d_loss: Discriminator loss Tensor
    :param g_loss: Generator loss Tensor
    :param learning_rate: Learning Rate Placeholder
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :return: A tuple of (discriminator training operation, generator training operation)
    """
    t_vars = tf.trainable_variables()
    g_vars = [var for var in t_vars if var.name.startswith('generator')]
    d_vars = [var for var in t_vars if var.name.startswith('discriminator')]
    all_update_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS)
    
    g_update_ops = [var for var in all_update_ops if var.name.startswith('generator')]
    d_update_ops = [var for var in all_update_ops if var.name.startswith('discriminator')]

    with tf.control_dependencies(d_update_ops):
        d_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(d_loss, var_list=d_vars)
    with tf.control_dependencies(g_update_ops):
        g_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(g_loss, var_list=g_vars)
    
    return d_train_opt, g_train_opt

"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_opt(model_opt, tf)
Tests Passed

Neural Network Training

Show Output

Use this function to show the current output of the generator during training. It will help you determine how well the GANs is training.

In [12]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import numpy as np

def show_generator_output(sess, n_images, input_z, out_channel_dim, image_mode):
    """
    Show example output for the generator
    :param sess: TensorFlow session
    :param n_images: Number of Images to display
    :param input_z: Input Z Tensor
    :param out_channel_dim: The number of channels in the output image
    :param image_mode: The mode to use for images ("RGB" or "L")
    """
    cmap = None if image_mode == 'RGB' else 'gray'
    z_dim = input_z.get_shape().as_list()[-1]
    example_z = np.random.uniform(-1, 1, size=[n_images, z_dim])

    samples = sess.run(
        generator(input_z, out_channel_dim, False),
        feed_dict={input_z: example_z})

    images_grid = helper.images_square_grid(samples, image_mode)
    pyplot.imshow(images_grid, cmap=cmap)
    pyplot.show()

Train

Implement train to build and train the GANs. Use the following functions you implemented:

  • model_inputs(image_width, image_height, image_channels, z_dim)
  • model_loss(input_real, input_z, out_channel_dim)
  • model_opt(d_loss, g_loss, learning_rate, beta1)

Use the show_generator_output to show generator output while you train. Running show_generator_output for every batch will drastically increase training time and increase the size of the notebook. It's recommended to print the generator output every 100 batches.

In [13]:
def train(epoch_count, batch_size, z_dim, learning_rate, beta1, get_batches, data_shape, data_image_mode):
    """
    Train the GAN
    :param epoch_count: Number of epochs
    :param batch_size: Batch Size
    :param z_dim: Z dimension
    :param learning_rate: Learning Rate
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :param get_batches: Function to get batches
    :param data_shape: Shape of the data
    :param data_image_mode: The image mode to use for images ("RGB" or "L")
    """
    inputs_real, inputs_z, lr = model_inputs(data_shape[1], data_shape[2], data_shape[3], z_dim)
    d_loss, g_loss = model_loss(inputs_real, inputs_z, data_shape[-1])
    d_train_opt, g_train_opt = model_opt(d_loss, g_loss, learning_rate, beta1)
    step = 0 
    
    with tf.Session() as sess:
        sess.run(tf.global_variables_initializer())
        for e in range(epoch_count):
            for batch_images in get_batches(batch_size):
    
                step += 1
                batch_images = batch_images * 2
                
                batch_z = np.random.uniform(-1, 1, size=(batch_size, z_dim))
                
                _ = sess.run(d_train_opt, feed_dict={inputs_real: batch_images, 
                                                     inputs_z: batch_z, 
                                                     lr:learning_rate})
                _ = sess.run(g_train_opt, feed_dict={inputs_z: batch_z, lr:learning_rate})
                
                if step % 100 == 0:
                    train_loss_d = d_loss.eval({inputs_z: batch_z, inputs_real: batch_images})
                    train_loss_g = g_loss.eval({inputs_z: batch_z})
                    print("Epoch {}/{} Step {}...".format(e+1, epoch_count, step),
                      "Discriminator Loss: {:.4f}...".format(train_loss_d),
                      "Generator Loss: {:.4f}".format(train_loss_g))    
                    show_generator_output(sess, 25, inputs_z, data_shape[3], data_image_mode)               

MNIST

Test your GANs architecture on MNIST. After 2 epochs, the GANs should be able to generate images that look like handwritten digits. Make sure the loss of the generator is lower than the loss of the discriminator or close to 0.

In [14]:
batch_size = 32
z_dim = 32
learning_rate = 0.0001
beta1 = 0.5


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 2

mnist_dataset = helper.Dataset('mnist', glob(os.path.join(data_dir, 'mnist/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, mnist_dataset.get_batches,
          mnist_dataset.shape, mnist_dataset.image_mode)
Epoch 1/2 Step 100... Discriminator Loss: 0.4874... Generator Loss: 2.4266
Epoch 1/2 Step 200... Discriminator Loss: 0.9700... Generator Loss: 1.0067
Epoch 1/2 Step 300... Discriminator Loss: 0.9656... Generator Loss: 0.8587
Epoch 1/2 Step 400... Discriminator Loss: 0.7693... Generator Loss: 1.9481
Epoch 1/2 Step 500... Discriminator Loss: 0.7000... Generator Loss: 2.4319
Epoch 1/2 Step 600... Discriminator Loss: 0.7323... Generator Loss: 1.5480
Epoch 1/2 Step 700... Discriminator Loss: 1.0564... Generator Loss: 0.7957
Epoch 1/2 Step 800... Discriminator Loss: 0.6428... Generator Loss: 1.7355
Epoch 1/2 Step 900... Discriminator Loss: 1.0828... Generator Loss: 0.7970
Epoch 1/2 Step 1000... Discriminator Loss: 0.9208... Generator Loss: 1.0599
Epoch 1/2 Step 1100... Discriminator Loss: 0.9086... Generator Loss: 1.6272
Epoch 1/2 Step 1200... Discriminator Loss: 1.0254... Generator Loss: 0.8482
Epoch 1/2 Step 1300... Discriminator Loss: 0.8470... Generator Loss: 1.5134
Epoch 1/2 Step 1400... Discriminator Loss: 0.7790... Generator Loss: 1.4836
Epoch 1/2 Step 1500... Discriminator Loss: 0.8681... Generator Loss: 1.4855
Epoch 1/2 Step 1600... Discriminator Loss: 0.7254... Generator Loss: 1.4066
Epoch 1/2 Step 1700... Discriminator Loss: 0.7911... Generator Loss: 1.4035
Epoch 1/2 Step 1800... Discriminator Loss: 0.7525... Generator Loss: 1.3710
Epoch 2/2 Step 1900... Discriminator Loss: 0.8410... Generator Loss: 1.0821
Epoch 2/2 Step 2000... Discriminator Loss: 1.0579... Generator Loss: 0.7984
Epoch 2/2 Step 2100... Discriminator Loss: 0.7056... Generator Loss: 1.4816
Epoch 2/2 Step 2200... Discriminator Loss: 0.9009... Generator Loss: 1.0031
Epoch 2/2 Step 2300... Discriminator Loss: 1.0486... Generator Loss: 0.8504
Epoch 2/2 Step 2400... Discriminator Loss: 1.2483... Generator Loss: 0.5985
Epoch 2/2 Step 2500... Discriminator Loss: 0.7176... Generator Loss: 1.2909
Epoch 2/2 Step 2600... Discriminator Loss: 0.7179... Generator Loss: 1.3468
Epoch 2/2 Step 2700... Discriminator Loss: 1.2335... Generator Loss: 0.5757
Epoch 2/2 Step 2800... Discriminator Loss: 0.7448... Generator Loss: 1.3028
Epoch 2/2 Step 2900... Discriminator Loss: 0.8103... Generator Loss: 1.1024
Epoch 2/2 Step 3000... Discriminator Loss: 0.8792... Generator Loss: 1.0145
Epoch 2/2 Step 3100... Discriminator Loss: 1.0179... Generator Loss: 0.8453
Epoch 2/2 Step 3200... Discriminator Loss: 0.8420... Generator Loss: 1.1676
Epoch 2/2 Step 3300... Discriminator Loss: 0.6635... Generator Loss: 1.7500
Epoch 2/2 Step 3400... Discriminator Loss: 0.6838... Generator Loss: 1.8746
Epoch 2/2 Step 3500... Discriminator Loss: 0.6479... Generator Loss: 1.4686
Epoch 2/2 Step 3600... Discriminator Loss: 0.5850... Generator Loss: 1.9922
Epoch 2/2 Step 3700... Discriminator Loss: 0.6085... Generator Loss: 1.9882

CelebA

Run your GANs on CelebA. It will take around 20 minutes on the average GPU to run one epoch. You can run the whole epoch or stop when it starts to generate realistic faces.

In [15]:
batch_size = 32
z_dim = 32
learning_rate = 0.0001
beta1 = 0.5


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 1

celeba_dataset = helper.Dataset('celeba', glob(os.path.join(data_dir, 'img_align_celeba/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, celeba_dataset.get_batches,
          celeba_dataset.shape, celeba_dataset.image_mode)
Epoch 1/1 Step 100... Discriminator Loss: 0.4911... Generator Loss: 3.0641
Epoch 1/1 Step 200... Discriminator Loss: 0.5400... Generator Loss: 2.7944
Epoch 1/1 Step 300... Discriminator Loss: 0.8014... Generator Loss: 2.6951
Epoch 1/1 Step 400... Discriminator Loss: 0.7207... Generator Loss: 1.8197
Epoch 1/1 Step 500... Discriminator Loss: 1.0179... Generator Loss: 1.3417
Epoch 1/1 Step 600... Discriminator Loss: 0.9508... Generator Loss: 1.8543
Epoch 1/1 Step 700... Discriminator Loss: 1.4690... Generator Loss: 0.7025
Epoch 1/1 Step 800... Discriminator Loss: 1.0450... Generator Loss: 1.3810
Epoch 1/1 Step 900... Discriminator Loss: 1.2371... Generator Loss: 0.9142
Epoch 1/1 Step 1000... Discriminator Loss: 1.0765... Generator Loss: 1.0187
Epoch 1/1 Step 1100... Discriminator Loss: 1.1913... Generator Loss: 0.7782
Epoch 1/1 Step 1200... Discriminator Loss: 1.0943... Generator Loss: 1.2589
Epoch 1/1 Step 1300... Discriminator Loss: 1.3461... Generator Loss: 0.7219
Epoch 1/1 Step 1400... Discriminator Loss: 1.2548... Generator Loss: 0.7795
Epoch 1/1 Step 1500... Discriminator Loss: 1.1513... Generator Loss: 1.1954
Epoch 1/1 Step 1600... Discriminator Loss: 0.9499... Generator Loss: 1.2960
Epoch 1/1 Step 1700... Discriminator Loss: 1.1842... Generator Loss: 1.0835
Epoch 1/1 Step 1800... Discriminator Loss: 1.1889... Generator Loss: 1.1334
Epoch 1/1 Step 1900... Discriminator Loss: 1.0134... Generator Loss: 0.9349
Epoch 1/1 Step 2000... Discriminator Loss: 1.0730... Generator Loss: 1.0894
Epoch 1/1 Step 2100... Discriminator Loss: 1.1101... Generator Loss: 0.8924
Epoch 1/1 Step 2200... Discriminator Loss: 1.3121... Generator Loss: 1.0275
Epoch 1/1 Step 2300... Discriminator Loss: 1.1012... Generator Loss: 1.1197
Epoch 1/1 Step 2400... Discriminator Loss: 1.1499... Generator Loss: 1.3728
Epoch 1/1 Step 2500... Discriminator Loss: 1.0231... Generator Loss: 1.3209
Epoch 1/1 Step 2600... Discriminator Loss: 1.1346... Generator Loss: 0.9654
Epoch 1/1 Step 2700... Discriminator Loss: 1.1728... Generator Loss: 0.9662
Epoch 1/1 Step 2800... Discriminator Loss: 1.1410... Generator Loss: 0.9947
Epoch 1/1 Step 2900... Discriminator Loss: 0.9673... Generator Loss: 1.1416
Epoch 1/1 Step 3000... Discriminator Loss: 1.1291... Generator Loss: 1.0416
Epoch 1/1 Step 3100... Discriminator Loss: 1.1063... Generator Loss: 0.9791
Epoch 1/1 Step 3200... Discriminator Loss: 1.1977... Generator Loss: 1.1651
Epoch 1/1 Step 3300... Discriminator Loss: 1.2145... Generator Loss: 0.9082
Epoch 1/1 Step 3400... Discriminator Loss: 1.0070... Generator Loss: 1.3746
Epoch 1/1 Step 3500... Discriminator Loss: 1.1118... Generator Loss: 0.9117
Epoch 1/1 Step 3600... Discriminator Loss: 1.1589... Generator Loss: 1.0281
Epoch 1/1 Step 3700... Discriminator Loss: 1.4428... Generator Loss: 0.6961
Epoch 1/1 Step 3800... Discriminator Loss: 1.2164... Generator Loss: 0.9007
Epoch 1/1 Step 3900... Discriminator Loss: 0.8601... Generator Loss: 1.2034
Epoch 1/1 Step 4000... Discriminator Loss: 1.1045... Generator Loss: 0.9673
Epoch 1/1 Step 4100... Discriminator Loss: 1.1018... Generator Loss: 0.9495
Epoch 1/1 Step 4200... Discriminator Loss: 0.8637... Generator Loss: 1.6139
Epoch 1/1 Step 4300... Discriminator Loss: 1.0557... Generator Loss: 1.0430
Epoch 1/1 Step 4400... Discriminator Loss: 1.1535... Generator Loss: 0.9485
Epoch 1/1 Step 4500... Discriminator Loss: 1.0684... Generator Loss: 1.0868
Epoch 1/1 Step 4600... Discriminator Loss: 0.8673... Generator Loss: 1.3671
Epoch 1/1 Step 4700... Discriminator Loss: 1.0625... Generator Loss: 1.0339
Epoch 1/1 Step 4800... Discriminator Loss: 0.9612... Generator Loss: 1.2300
Epoch 1/1 Step 4900... Discriminator Loss: 1.0857... Generator Loss: 1.0022
Epoch 1/1 Step 5000... Discriminator Loss: 1.2997... Generator Loss: 0.8343
Epoch 1/1 Step 5100... Discriminator Loss: 0.9936... Generator Loss: 1.0598
Epoch 1/1 Step 5200... Discriminator Loss: 1.1287... Generator Loss: 0.8938
Epoch 1/1 Step 5300... Discriminator Loss: 1.1481... Generator Loss: 0.7899
Epoch 1/1 Step 5400... Discriminator Loss: 0.9575... Generator Loss: 1.4204
Epoch 1/1 Step 5500... Discriminator Loss: 1.0402... Generator Loss: 1.1204
Epoch 1/1 Step 5600... Discriminator Loss: 0.9829... Generator Loss: 1.1365
Epoch 1/1 Step 5700... Discriminator Loss: 1.0209... Generator Loss: 1.0170
Epoch 1/1 Step 5800... Discriminator Loss: 1.0116... Generator Loss: 0.9596
Epoch 1/1 Step 5900... Discriminator Loss: 0.9346... Generator Loss: 1.3141
Epoch 1/1 Step 6000... Discriminator Loss: 1.2304... Generator Loss: 0.9969
Epoch 1/1 Step 6100... Discriminator Loss: 1.1744... Generator Loss: 0.9393
Epoch 1/1 Step 6200... Discriminator Loss: 0.9822... Generator Loss: 1.2234
Epoch 1/1 Step 6300... Discriminator Loss: 1.1290... Generator Loss: 1.2507

Submitting This Project

When submitting this project, make sure to run all the cells before saving the notebook. Save the notebook file as "dlnd_face_generation.ipynb" and save it as a HTML file under "File" -> "Download as". Include the "helper.py" and "problem_unittests.py" files in your submission.